---
title: "a16z’s Olivia Moore: 90% of Generative AI Apps Grow via Organic Traffic"
author: "Unique Research"
sourcePublication: "Unique Research (FFCAP English library)"
originalPublishedAt: "2023-10-12T19:30:00+08:00"
canonical: "https://ffcap.cn/en/research/20231012a08wu000"
source: "https://ffcap.cn/en/research/20231012a08wu000"
language: "en"
cnSourceUrl: "https://view.inews.qq.com/a/20231012A08WU000"
---

# a16z’s Olivia Moore: 90% of Generative AI Apps Grow via Organic Traffic

_Unique Research · Ranking / a16z consumer genAI · October 12, 2023 · KEEP type ranking — ~3,140 CJK / Top-50 traffic findings preserved_

Olivia Moore (a16z consumer investing partner; ex-CRV, Goldman) — [How Are Consumers Using Generative AI?](https://a16z.com/how-are-consumers-using-generative-ai/). Nine months after ChatGPT; seven months after it became the fastest consumer app to 100M MAU. Method: SimilarWeb Top-50 generative-AI **web** apps by monthly visits (desktop+mobile web, data through Jun 2023); for apps with mobile clients, Sensor Tower “traffic” ≈ MAUs × sessions/MAU/month folded into ranking. Excludes app-only companies and Discord-only activity. Ranking is a trend tool, not an exhaustive AI-platform census.

## Core takeaways

1.  **GenAI apps rose fast from a blank sheet** — most leading apps built from scratch; **80%** of sites did not exist a year earlier. Traditional companies are AI-augmenting, but the stickiest UX is often new.
2.  **ChatGPT’s lead** — ~**60%** of Top-50 monthly traffic; ~global #24 site. Still small vs WhatsApp/YouTube/Facebook.
3.  **LLM assistants dominate** (~68% of Top-50 user traffic) — companions and content tools rising (image gen 41% of content category traffic; pro writing 26%; video 8%).
4.  **Early winners, most categories still contestable** — few 1st–2nd gaps >2×; average ~50% MoM growth over prior 6 months.
5.  **Organic acquisition + willingness to pay** — most have little paid; **90%** already generating revenue, nearly all subscription; avg ~$21/mo (~$252/yr).
6.  **Mobile still emerging** — browser-first so far; more mobile-first expected as stacks mature.

## 1\. Built from scratch around genAI

Only 5 of Top 50 are big-tech apps or acquisitions: Bard (Google), Poe (Quora), QuillBot (Course Hero), Pixlr (123RF), Clipchamp (Microsoft). PitchBook: up to **48%** of the rest bootstrapped with no outside capital; another **15%** raised >$50M. Stack split among Top 50 roughly thirds: train proprietary / fine-tune / UI wrappers — but among Top 10, half own models, 4 fine-tune, only 1 wrapper. Ex-ChatGPT (OpenAI’s ~$11.3B skews averages): proprietary-model cos avg raise ~$98M; open-model ~$20M; wrappers ~$9M.

## 2\. ChatGPT leads

~**1.6B** monthly visits / ~**200M** monthly users (Jun 2023). #2 Character.AI ~21% of ChatGPT’s scale — on mobile, Sensor Tower DAUs comparable to ChatGPT with higher retention. Combined web+app, ChatGPT sits near Reddit/LinkedIn/Twitch — far below mega platforms.

## 3\. Assistants dominate; companions & creative rise

Beyond ChatGPT: Bard and Poe both Top 5. Model hubs punch above count: Civitai (images) and Hugging Face both Top 10 — and web traffic understates local downloads.

## 4\. Categories still open

Most 1st–2nd traffic gaps <2×. Specialization alongside general tools: Midjourney leads broad image gen while Leonardo (game assets) grows millions of MUV alongside Midjourney’s Discord growth.

## 5\. Healthy acquisition; users pay

GenAI flipped the paid-UA treadmill. Bottom-quartile genAI apps: only ~**2%** paid traffic vs ~**70%** for a16z’s non-AI consumer-subscription benchmark of 150 apps. Growth via X, Reddit, Discord, email, word of mouth. ARPU (~$21/mo) above typical Calm/Headspace/Duolingo bands (<$70/yr; ~$10/mo).

## 6\. Mobile still new

ChatGPT took ~6 months to ship mobile. Image-gen APIs let Lensa/WOMBO spike then fade. Only ~15 Top-50 have mobile apps; almost all get <10% of traffic from apps vs web. Exceptions: PhotoRoom (~88% app), Character.AI (~46%), Speechify (~20%) — higher sessions/visitor on mobile. Users spend ~36 more minutes/day on mobile than desktop (4.1h vs 3.5h) — expect more mobile-first genAI.

---

Editorial English edition published on UniqueCapital: https://ffcap.cn/en/research/20231012a08wu000
On-site reading page: https://ffcap.cn/en/research/20231012a08wu000